عنوان مقاله [English]
Background and Objectives
Drought stress is one of the most substantial environmental stresses affecting agricultural productivity around the world and may result in considerable yield reduction. The human population is expected to increase to over 8 billion by the year 2020. Therefore, providing adequate food and preventing crop yield reduction is essential. Plant breeding can solve this problem to some extent. Evaluating yield components and their inter-relationships as well as detecting suitable selection indices are very important in the safflower breeding program. The multivariate statistical analysis can provide more insights into the deep structure of data and traits’ relationships
Materials and Methods
In order to study the relationships between yield components of eight genotypes of safflowerunder normal irrigationand drought stress conditions and determine the high yield genotypes, a split-plot experiment was conducted in a randomized complete block design with three replications in the research field of Imam Khomeini International University, Qazvin. Irrigation regime included normal irrigation(irrigation after 60 mm evaporation from class A pan evaporation) and drought stress (watering until 50% flowering similar to normal irrigationand 50% flowering to maturity and irrigation after 100 mm evaporation from class A pan evaporation) were considered as main plots. Eight genotypes, including Kuseh Local, Sina, Isfahan Local, Mexican 88, Faraman, Soffeh, Goldasht, and Mexican 11 were used too.
The analysis of the variance showed that the irrigation effect was significant for traits 1000 grain weight, flower weight, harvest index, and grain yield. There were significant differences among the genotypes for most of the measured traits, except boll diameter, boll number per plant, relative water content, and relative water loss. Plant height, shoot number, boll number, stem diameter, chlorophyll value, flower weight, biological yield, and harvest index positively correlated with the grain yield in both normal irrigation and drought stress conditions. Based on the results of stepwise regression and path analysis, biological yield and harvest index in both normal irrigation and drought stress conditions had the greatest effect on grain yield. Due to the high correlation between yield and harvest index and biological yield under stress conditions, the selection of varieties with high harvest index and biological yield can be very effective in achieving maximum yield under stress conditions. Factor analysis revealed that 4 factors accounted for approximately 82% variance changes in both normal irrigation and drought stress conditions. These factors were called yield, seed, boll size, and leaf water, respectively under normal irrigation, andplant vigor, yield, seed, boll size, and leaf water, respectively under drought stress conditions. Based on the seed yield factor and the biplot display, Faraman, Mahali Isfahan, Sina, and Mahali Kuseh were determined as high yield genotypes both in normal and drought stress conditions.
Grain yield is a quantitative trait with a low broad sense heritability. Therefore, identifying the traits correlated with grain yield is a suitable strategy for the improvement and the indirect selection of grain yield. Stepwise regression and path analysis are efficient and useful statistical approaches for reaching the above aim. According to the results, the traits of the harvest index and biomass were recognized as the best suitable traits for indirect grain yield selection under normal and drought stress conditions.